Merge branch 'release/0.6.3'

Author: emfomy

.pylintrc

@@ -3,5 +3,6 @@
 max-line-length = 128
 
 disable =
-	missing-docstring,
+	bad-continuation,
 	duplicate-code,
+	missing-docstring,

Makefile

@@ -22,6 +22,7 @@ doc:
 	( cd docs ; make clean ; make html )
 
 upload: dist check
+	ls dist/*.tar.gz
 	$(TWINE) upload --repository-url https://test.pypi.org/legacy/ dist/*.tar.gz --verbose
 
 clean:

README.rst

@@ -73,17 +73,18 @@ Requirements
 
 * `Python <http://www.python.org>`_ 3.5+
 * `Cython <http://cython.org>`_ 0.29+
+* `TreeLib <https://treelib.readthedocs.io>`_ 1.5+
 
 .. note::
    For Python 2 users, please use `PyCkip 0.4.2 <https://pypi.org/project/pyckip/0.4.2/>`_ instead.
 
-CkipWs (Optional)
+CKIPWS (Optional)
 ^^^^^^^^^^^^^^^^^
 
 * `CKIP Word Segmentation <http://ckip.iis.sinica.edu.tw/project/wordsegment/>`_ Linux version 20190524+
 
-CkipParser (Optional)
-^^^^^^^^^^^^^^^^^^^^^
+CKIP-Parser (Optional)
+^^^^^^^^^^^^^^^^^^^^^^
 
 * `CKIP Parser <http://ckip.iis.sinica.edu.tw/project/parser/>`_ Linux version 20190506+ (20190725+ recommended)
 
@@ -191,8 +192,15 @@ Utilities
 
    # Format CkipWs output
    ws_text = ['中文字(Na)　喔(T)', '啊哈(I)　哈哈(D)']
-   for text in ws_text: print(ckipnlp.util.ws.WsSentence.from_text(text))
-   for text in ws_text: print(repr(ckipnlp.util.ws.WsSentence.from_text(text)))
+
+   # Show Sentence List
+   ws_sents = WsSentenceList.from_text(ws_text)
+   print(repr(ws_sents))
+   print(ws_sents.to_text())
+
+   # Show Each Sentence
+   for ws_sent in ws_sents: print(repr(ws_sent))
+   for ws_sent in ws_sents: print(ws_sent.to_text())
 
    # Show CkipParser output as tree
    tree_text = 'S(theme:NP(property:N‧的(head:Nhaa:我|Head:DE:的)|Head:Nad(DUMMY1:Nab:早餐|Head:Caa:和|DUMMY2:Naa:午餐))|quantity:Dab:都|Head:VC31:吃完|aspect:Di:了)'
@@ -206,31 +214,39 @@ Utilities
    for node in tree.get_heads(1): print(node)
 
    # Get relations
-   for r in tree.get_relations(0): print(r)
+   for rel in tree.get_relations(0): print(rel)
 
 
 FAQ
 ===
 
+.. warning::
+
+   Due to C code implementation, one should not instance more than one ``CkipWs`` driver object and one ``CkipParser`` driver object.
+
+------------
+
 .. warning::
 
    The CKIPWS throws "``what():  locale::facet::_S_create_c_locale name not valid``". What should I do?
 
-Install locale data.
+   Install locale data.
 
-.. code-block:: bash
+   .. code-block:: bash
 
-   apt-get install locales-all
+      apt-get install locales-all
+
+------------
 
 .. warning::
 
    The CKIPParser throws "``ImportError: libCKIPParser.so: cannot open shared object file: No such file or directory``". What should I do?
 
-Add below command to ``~/.bashrc``:
+   Add below command to ``~/.bashrc``:
 
-.. code-block:: bash
+   .. code-block:: bash
 
-   export LD_LIBRARY_PATH=<ckipparser-linux-root>/lib:$LD_LIBRARY_PATH
+      export LD_LIBRARY_PATH=<ckipparser-linux-root>/lib:$LD_LIBRARY_PATH
 
 License
 =======

ckipnlp/__init__.py

@@ -6,7 +6,7 @@
 __copyright__ = '2018-2019 CKIP Lab'
 
 __title__ = 'CKIPNLP'
-__version__ = '0.6.2'
+__version__ = '0.6.3'
 __description__ = 'CKIP CoreNLP Wrappers'
 __license__ = 'CC BY-NC-SA 4.0'
 

ckipnlp/parser/__init__.py

@@ -27,17 +27,30 @@ class CkipParser:
 
     Parameters
     ----------
-    logger : bool
-        enable logger.
-    inifile : str
-        the path to the INI file.
-    wsinifile : str
-        the path to the INI file for CKIPWS.
-    options:
-        the options, see :func:`ckipnlp.util.ini.create_ws_ini` and :func:`ckipnlp.util.ini.create_parser_ini`
+        logger : bool
+            enable logger.
+        inifile : str
+            the path to the INI file.
+        wsinifile : str
+            the path to the INI file for CKIPWS.
+    Other Parameters
+    ----------------
+        **
+            the configs for CKIPParser, ignored if **inifile** is set. Please refer :func:`ckipnlp.util.ini.create_parser_ini`.
+        **
+            the configs for CKIPWS, ignored if **wsinifile** is set. Please refer :func:`ckipnlp.util.ini.create_ws_ini`.
+
+    Warning
+    -------
+        Never instance more than one object of this class!
     """
 
-    def __init__(self, *, logger=False, inifile=None, wsinifile=None, **options):
+    def __init__(self, *,
+        logger=False,
+        inifile=None,
+        wsinifile=None,
+        **kwargs,
+    ):
 
         self.__core = CkipParserCore()
 
@@ -48,19 +61,19 @@ def __init__(self, *, logger=False, inifile=None, wsinifile=None, **options):
         if not wsinifile:
             fwsini = _tempfile.NamedTemporaryFile(mode='w')
             wsinifile = fwsini.name
-            wsinidata, options = create_ws_ini(**options)
+            wsinidata, kwargs = create_ws_ini(**kwargs)
             fwsini.write(wsinidata)
             fwsini.flush()
 
         if not inifile:
             fini = _tempfile.NamedTemporaryFile(mode='w')
             inifile = fini.name
-            inidata, options = create_parser_ini(wsinifile=wsinifile, **options)
+            inidata, kwargs = create_parser_ini(wsinifile=wsinifile, **kwargs)
             fini.write(inidata)
             fini.flush()
 
         def CkipParser(*, _=None): pass # pylint: disable=redefined-outer-name, invalid-name, multiple-statements
-        CkipParser(**options)
+        CkipParser(**kwargs)
 
         self.__core.init_data(inifile)
 
@@ -82,17 +95,16 @@ def apply(self, text):
 
         Parameters
         ----------
-        text : str
-            the input sentence.
+            text : str
+                the input sentence.
 
         Return
         ------
-        str
-            the output sentence.
+            str
+                the output sentence.
 
-        Notes
-        -----
-        One may also call this method as :func:`__call__`.
+        .. note::
+            One may also call this method as :func:`__call__`.
         """
         return self.apply_list([text])[0]
 
@@ -101,13 +113,13 @@ def apply_list(self, ilist):
 
         Parameters
         ----------
-        ilist: list
-            the list of input sentences (str).
+            ilist
+                the list of input sentences.
 
         Return
         ------
-        olist: list
-            the list of output sentences (str).
+            List[str]
+                the list of output sentences.
         """
         return self.__core.apply_list(ilist)
 
@@ -116,9 +128,9 @@ def apply_file(self, ifile, ofile):
 
         Parameters
         ----------
-        ifile: str
-             the input file.
-        ofile: str
-             the output file (will be overwritten).
+            ifile : str
+                 the input file.
+            ofile : str
+                 the output file (will be overwritten).
         """
         return self.__core.apply_file(ifile, ofile)

ckipnlp/util/ini.py

@@ -12,31 +12,30 @@
 
 import ckipnlp as _about
 
-def create_ws_ini( # pylint: disable=too-many-statements
-        *,
-        data2dir=None,
-        lexfile=None,
-        new_style_format=False,
-        show_category=True,
-        sentence_max_word_num=80,
-        **options,
+def create_ws_ini(*, # pylint: disable=too-many-statements
+    data2dir=None,
+    lexfile=None,
+    new_style_format=False,
+    show_category=True,
+    sentence_max_word_num=80,
+    **options,
 ):
     """Generate CKIP word segmentation config.
 
     Parameters
     ----------
-    data2dir : str
-        the path to the folder "Data2/".
-    lexfile : str
-        the path to the user-defined lexicon file.
-
-    new_style_format : bool
-        split sentences by newline characters ("\\\\n") rather than punctuations.
-    show_category : bool
-        show part-of-speech tags.
-
-    sentence_max_word_num : int
-        maximum number of words per sentence.
+        data2dir : str
+            the path to the folder "Data2/".
+        lexfile : str
+            the path to the user-defined lexicon file.
+
+        new_style_format : bool
+            split sentences by newline characters ("\\\\n") rather than punctuations.
+        show_category : bool
+            show part-of-speech tags.
+
+        sentence_max_word_num : int
+            maximum number of words per sentence.
     """
 
     # pylint: disable=invalid-name
@@ -312,35 +311,34 @@ def create_ws_ini( # pylint: disable=too-many-statements
 
     return '\n'.join(cfg), options
 
-def create_parser_ini( # pylint: disable=too-many-statements
-        *,
-        wsinifile,
-        ruledir=None,
-        rdbdir=None,
-        do_ws=True,
-        do_parse=True,
-        do_role=True,
-        sentence_delim='，,；。！？',
-        **options,
+def create_parser_ini(*, # pylint: disable=too-many-statements
+    wsinifile,
+    ruledir=None,
+    rdbdir=None,
+    do_ws=True,
+    do_parse=True,
+    do_role=True,
+    sentence_delim='，,；。！？',
+    **options,
 ):
     """Generate CKIP parser config.
 
     Parameters
     ----------
-    ruledir : str
-        the path to "Rule/".
-    rdbdir : str
-        the path to "RDB/".
-
-    do_ws : bool
-        do word-segmentation.
-    do_parse : bool
-        do parsing.
-    do_role : bool
-        do role.
-
-    sentence_delim: str
-        the sentence delimiters.
+        ruledir : str
+            the path to "Rule/".
+        rdbdir : str
+            the path to "RDB/".
+
+        do_ws : bool
+            do word-segmentation.
+        do_parse : bool
+            do parsing.
+        do_role : bool
+            do role.
+
+        sentence_delim : str
+            the sentence delimiters.
     """
 
     # pylint: disable=invalid-name